2. EPIDURAL AND OPIOD EFFECTS ON FEVER 219
consent, we studied eight healthy male volunteers, each if necessary, oxygen was given via nasal cannula to
on four study days. None was obese, was taking medi- maintain oxygen saturation measured by pulse oximetry
cation, or had a history of thyroid disease or Raynaud (SpO2) more than 95%. Fentanyl was infused for 8 h and
syndrome. Morphometric and demographic characteris- then discontinued.
tics included: age, 27 6 yr; height, 180 8 cm;
weight, 74 8 kg; and body fat, 17 5%.
Measurements
The volunteers fasted for 8 h before arriving at the
Core temperature was measured at the tympanic mem-
laboratory. They were minimally clothed and reclined on
a standard operation table during the study. Ambient brane using Mon-a-Therm thermocouples (Mallinckrodt
temperature was maintained near 22°C and relative hu- Anesthesiology Product, Inc., St. Louis, MO). Mean skin-
midity near 45%. To avoid circadian fluctuations, studies surface temperature was calculated from measurements
were scheduled at the same time each day. at 15 area-weighted sites.15 Temperatures were recorded
An 18-gauge catheter was inserted in a left forearm at every 5 min from thermocouples connected to Iso-
vein for fluid and drug administration. Lactated Ringer’s Thermex thermometers having an accuracy of 0.1°C and
solution at ambient temperature was infused at a rate of a precision of 0.01°C (Columbus Instruments Corp., Co-
100 ml/h. A 16-gauge catheter was inserted in the right lumbus, OH).
median-cephalic vein for blood sampling. Pupillary responses were measured to evaluate the
pharmacodynamic effect of fentanyl. An infrared pupil-
lometer (Fairville Medical Optics, Buckinghamshire, UK)
Protocol was programmed to provide a 0.5-s, 130 candela/m2
On each study day, the volunteers were randomly pulse of green light and to scan the pupil at the rate of
assigned to either: (1) a control day when no opioid or 10 Hz for 2 s from the beginning of the light stimulus.
epidural analgesia was given; (2) epidural analgesia using We have previously described this method16 and used it
0.2% ropivacaine alone; (3) epidural analgesia using 0.2%
to quantify opioid effect.17 The maximum reduction in
ropivacaine in combination with 2.0 g/ml fentanyl; or
pupil size during the 2-s scan identified the reflex ampli-
(4) intravenous fentanyl at a target plasma concentration
tude. Ambient light was maintained near 150 lux, and
of 2.5 ng/ml. Volunteers were then given an intravenous
the contralateral eye was covered during measurements.
injection of 50 IU/g of human recombinant IL-2 (elapsed
Peripheral venous blood for fentanyl analysis was sam-
time, 0), observed 2 h later by 100 IU/g of the drug
pled just before IL-2 administration and at 1, 3, 5, and 7
(Chiron, Inc., Berkeley, CA). At least 2 weeks were
elapsed h. The plasma samples were stored at 20°C
allowed between the fentanyl day and the combined
until analysis by gas chromatography, using techniques
epidural ropivacaine–fentanyl day to minimize tolerance.
described by Bjorkman and Stanski18 and Selinger et al.19
On the epidural days, a catheter was inserted into the
The limit of detection is near 0.2 ng/ml.
epidural space via the L2–L3 interspace. The epidural
catheter was then injected with 2 ml lidocaine, 2%, with We evaluated IL-6, IL-8, and TNF , as well as the
epinephrine 1:100,000. This test dose was followed in antiinflammatory cytokine IL-10 hourly.20 –23 Plasma IL-6
5 min by 10 –12 ml of 0.2% ropivacaine (Ropivacaine and IL-8 concentrations were measured by an enzyme-
HCl; Astra, Inc., Westborough, MA) or ropivacaine in linked immunosorbent assay (Human interleukin-6 and
combination with 2 g/ml fentanyl without epineph- interleukin-8 ELISA Kits; Toray Industries, Inc., Tokyo,
rine. The initial anesthetic dose was based on the volun- Japan). TNF concentrations were determined by a
teers’ heights and calculated to produce a dermatomal human immunoassay (Quantikine HS; R&D Systems,
level of sensory blockade near T8 –S1 bilaterally as de- Minneapolis, MN). Plasma IL-10 concentrations were de-
termined by loss of cutaneous cold sensation and re- termined by a solid-phase enzyme-amplified immunoas-
sponse to pinprick. Epidural analgesia was then main- say (IL-10 EASIA Kit; Medgenix Diagnostics S.A., Fleurus,
tained with the continuously administered drug at a rate Belgium). In each case, the assays were performed per
of 8 –12 ml/h to maintain the target sensory block level. the manufacturers’ directions, and appropriate calibra-
On the intravenous fentanyl day, fentanyl was admin- tion curves were constructed. All are highly specific and
istered using a pump (Ohmeda 9000; Ohmeda, Steeton, sensitive over the range of observed values.
UK) programmed to target fentanyl blood concentra- Heart rates and arterial oxygen saturation were moni-
tions of 2.5 ng/ml using a modification of the Kruger– tored continuously using pulse oximetry (Biox 3700;
Thiemer method11 and published data.12 This is a low Ohmeda, Salt Lake City, UT). Blood pressure at the ankle
concentration that produces only mild sedation and is was determined oscillometrically (Critikon, Tampa, FL).
similar to concentrations that may be used for labor or End-tidal carbon dioxide concentrations and respiratory
postoperative pain.13,14 To maintain an appropriate end- rates were monitored using a Capnomac Ultima (Datex
tidal carbon dioxide tension (PCO2), the volunteers were Medical Instruments, Tewksbury, MA). All temperatures
reminded to breathe during fentanyl administration, and, and hemodynamic data were recorded at 5-min intervals.
Anesthesiology, V 94, No 2, Feb 2001
3. 220 NEGISHI ET AL.
Table 1. Plasma Fentanyl Concentration and Pupillary Responses
Elapsed Time (h)
0 1 3 5 7
Control
Pupil size (mm) 5.7 0.8 5.6 0.6 6.0 0.6 5.9 0.9 5.5 1.2
Reflex amplitude (mm) 2 0.4 2.2 0.5 2.2 0.3 2.2 0.5 2.2 0.7
Epidural ropivacaine
Pupil size (mm) 5.8 1.0 5.5 0.7 5.7 1.1 5.8 1.0 6.0 1.0
Reflex amplitude (mm) 2.0 0.5 1.9 0.3 2.1 0.6 2.5 0.9 2.2 0.4
Epidural ropivacaine and fentanyl
[Fentanyl] (ng/ml) 0.2 0.2 0.3 0.2 0.3 0.1 0.3 0.3 0.2 0.2
Pupil size (mm) 5.6 0.7 5.7 0.5 5.3 0.7 5.0 1.0 5.2 1.3
Reflex amplitude (mm) 1.9 0.4 2.0 0.4 1.9 0.5 2.2 0.4 2.0 0.5
Intravenous fentanyl
[Fentanyl] (ng/ml) 1.1 1.2 1.8 0.8 1.4 0.4 1.5 0.3 1.5 0.4
Pupil size (mm) 5.8 0.8 3.2 0.8* 3.0 0.9* 3.1 0.7* 2.9 0.7*
Reflex amplitude (mm) 2.0 0.4 1.2 0.7* 1.1 0.5* 1.1 0.4* 1.1 0.5*
* Statistically significant differences from elapsed time 0. Pupil sizes during intravenous fentanyl administration all differed significantly from the other treatment
days.
Data Analysis changes were similarly evaluated. A chi-square analysis
Febrile responses on each study day are presented as was used to evaluate the fraction of the temperature
time-dependent changes. Specifically, we considered in- measurements in each group exceeding 38.0, 38.5, and
tegrated core temperature, peak temperature, and the 39.0°C. Results are presented as mean SD; P 0.05
time-to-peak temperature. Values were integrated during was considered statistically significant.
3– 8 elapsed h, with respect to the mean temperature
during the first elapsed hour. That is, we calculated the
Results
area under the temperature curve. This quantified the
extent to which core temperature exceeded initial val- The volunteers were only mildly sedated by intrave-
ues (in °C h) and is a standard way of expressing fever nous fentanyl. Total plasma fentanyl concentrations during
magnitude. administration of fentanyl averaged 1.5 0.2 ng/ml on the
Ambient temperature and humidity, hemodynamic re- intravenous fentanyl day, but only 0.2 0.1 ng/ml when
sponses, end-tidal PCO2, respiratory rate, SpO2, and admin- fentanyl was given epidurally. Pupil size and reflex am-
istered fluid volume on each study day were first averaged plitude were significantly reduced by intravenous fenta-
within each volunteer during 3– 8 elapsed h; the result- nyl (table 1). There were no significant differences in
ing values were then averaged among volunteers. ambient temperature, relative humidity, heart rate,
Most results were compared using repeated-measures blood pressure, end-tidal PCO2, and SpO2 on the four
analysis of variance and Scheffé F test. Time-dependent study days.
Table 2. Environmental, Hemodynamic, Respiratory, and Thermoregulatory Data
Epidural Ropivacaine
Control Epidural Ropivacaine and Fentanyl Intravenous Fentanyl
Ambient temperature (°C) 21.9 0.7 21.7 0.2 21.7 0.4 21.4 0.4
Relative humidity (%) 47 5 48 5 47 4 48 3
Initial core temperature (°C) 36.6 0.2 36.6 0.2 36.6 0.2 36.7 0.3
Mean arterial blood pressure (mmHg) 88 9 79 8 80 10 89 11
Heart rate (beats/min) 92 12 90 13 87 13 92 13
End-tidal PCO2 (mmHg) 37 4 33 9* 35 8 43 4
Administered fluid volume (l) 1.2 0.2 1.2 0.2 1.1 0.2 1.1 0.2
Maximum core temperature (°C) 38.7 0.6* 38.7 0.7* 38.7 0.9* 38.1 0.7
Time to maximum core temperature (h) 6.1 1.0 6.6 0.9* 6.6 0.6* 5.8 0.9
Integrated core temperature (°C h) 7.0 3.2* 7.0 3.3* 6.0 3.8* 3.8 3.0
Mean skin temperature (°C) 34.9 0.5* 35.3 0.7* 34.7 0.8 33.8 1.1
All values were averaged or integrated over 3– 8 elapsed hours, corresponding to the period of epidural analgesia or intravenous fentanyl administration. Data
are presented as mean SD.
* Statistically significant differences from intravenous fentanyl. There were no significant differences among the other treatment groups.
PCO2 partial pressure of carbon dioxide.
Anesthesiology, V 94, No 2, Feb 2001
4. EPIDURAL AND OPIOD EFFECTS ON FEVER 221
Discussion
Epidural analgesia is associated with a high incidence
of hyperthermia,1,5,8 which provokes expensive and in-
vasive interventions.6,7 Because passive hyperthermia
and excessive heat production are unlikely causes, ele-
vated body temperature in laboring and postoperative
patients is presumably true fever resulting from infec-
tion, tissue damage atelectasis, and so forth. The conven-
tional understanding is that hyperthermia associated
with epidural analgesia in the same clinical situation is
caused by epidural analgesia. However, this theory is
inconsistent with our observation that epidural analgesia
failed to influence febrile responses. Instead, our data
suggest that fever is inhibited by opioid administration in
the “control” subjects.
Although maximum and average core temperatures
differed by less than 1°C when the volunteers were
given fentanyl and epidural analgesia, integrated temper-
Fig. 1. Core temperatures after administration of 50 IU/g of ature was halved by low-dose fentanyl. More impor-
interleukin (IL) 2 followed by a second dose of 100 IU/g, 2 h tantly, the fraction of core temperatures exceeding var-
later. The first dose of IL-2 defined elapsed time 0. Data are
presented as mean SD. Error bars for the epidural days are ious temperatures was reduced by a factor of two to five.
omitted for clarity, but were similar to those shown for control This is a critical outcome because temperatures exceed-
and intravenous fentanyl. See table 2 for statistical analysis. ing specific values often trigger laboratory investigations
for infection and even antibiotic administration. That
Core temperatures peaked at 38.7 0.6°C on the intravenous fentanyl reduces the fraction of tempera-
control day, at 38.7 0.7°C during epidural ropivacaine, tures exceeding various threshold temperatures thus ex-
at 38.7 0.9°C with combined epidural ropivacaine– plains the high incidence of “fever work-ups” in laboring
fentanyl, but only at 38.1 0.7°C during intravenous mothers and their children after epidural analgesia.6,7 It
fentanyl (table 2). Fentanyl significantly reduced the fe- is also consistent with the fact that there is no evidence
brile response to pyrogen, decreasing integrated core whatsoever suggesting that these additional infection
temperature from 7.0 3.2°C h on the control day to investigations were otherwise justified or in any way
3.8 3.0°C h on the intravenous fentanyl day. In influenced outcome.2
contrast, epidural ropivacaine and epidural ropivacaine– Published reports demonstrate a correlation between
fentanyl did not inhibit manifestation of fever (fig. 1). hyperthermia during epidural analgesia and clinical signs
The fraction of core-temperature measurements that of infection.24 There is also evidence that hyperthermia
exceeded 38°C was halved by intravenous fentanyl. The during epidural analgesia is highly associated with pla-
fraction of measurements exceeding 38.5°C was re- cental inflammation.25 These authors concluded: “Epi-
duced more than fivefold by intravenous fentanyl (table dural analgesia is associated with intrapartum fever, but
3). Both these reductions were statistically significant only in the presence of placental inflammation. This
(P 0.001). suggests that the fever reported with epidural analgesia
Plasma concentration of TNF , IL-6, and IL-8 increased results from immune responses rather than the analgesia
later, reaching peak concentrations after 4 or 5 elapsed itself.” These results are consistent with our theory that
h. All subsequently decreased to near-baseline values by hyperthermia during epidural analgesia is associated
8 or 9 elapsed h. In contrast, IL-10 continued to increase with inflammation (although not necessarily infection).
throughout the study period. None of the cytokine con- Our results thus suggest that hyperthermia during epi-
centrations differed significantly on the four study days. dural analgesia should be taken seriously and not con-
Table 3. Core Temperatures Exceeding 38.0, 38.5, and 39.0°C
Control Epidural Ropivacaine Epidural Ropivacaine and Fentanyl Intravenous Fentanyl P
38.0°C 72 68 69 28 0.001
38.5°C 31 46 39 6 0.001
39.0°C 16 17 20 2 0.059
The percentage of measurements, during the 3– 8 elapsed hour period, in which core temperatures exceeded 38.0, 38.5, or 39.0°C.
Anesthesiology, V 94, No 2, Feb 2001
5. 222 NEGISHI ET AL.
sidered a benign complication of the anesthetic tech- Epidural analgesia during labor and maternal fever. ANESTHESIOLOGY 1999; 90:
1271–5
nique per se. In contrast, threshold temperatures 3. Vinson DC, Thomas R, Kiser T: Association between epidural analgesia
triggering investigations for infection should probably be during labor and fever. J Fam Pract 1993; 36:617–22
4. Macaulay JH, Mrcog KB, Steer PJ: Epidural analgesia in labor and fetal
reduced by approximately 0.5°C in patients given opioid hyperthermia. Obstet Gynecol 1992; 80:665–9
analgesia. 5. Bredtmann RD, Herden HN, Teichmann W, Moecke HP, Kniesel B, Baetgen
R, Techklenburg A: Epidural analgesia in colonic surgery: Results of a randomized
Opioids are frequently administered epidurally be- prospective study. Br J Surg 1990; 77:638 – 42
cause they augment analgesia with little respiratory com- 6. Lieberman E, Lang JM, Frigoletto F, Richardson DK, Ringer SA, Cohen A:
Epidural analgesia, intrapartum fever, and neonatal sepsis evaluation. Pediatrics
promise.26 Relatively little opioid is required in the epi- 1997; 99:415–9
dural space compared with intravenous administration, 7. Lieberman E, Cohen A, Lang J, Frigoletto F, Goetzl L: Maternal intrapartum
temperature elevation as a risk factor for cesarean delivery and assisted vaginal
and the dose we used was typical.26,27 Plasma fentanyl delivery. Am J Public Health 1999; 89:506 –10
concentrations were six- to sevenfold lower after epi- 8. Frigoletto FD, Lieberman E, Lang JM, Cohen A, Barss V, Ringer S, Datta S: A
clinical trial of active management of labor. N Engl J Med 1995; 333:745–50
dural than intravenous administration of the drug, which 9. Negishi C, Kim J-S, Lenhardt R, Sessler DI, Ozaki M, Vuong K, Bastanmehr
is consistent with the relatively small epidural dose.28 H, Bjorksten AR: Alfentanil reduces the febrile response to interleukin-2 in men.
Crit Care Med 1999; 28:1295–300
Furthermore, pupil size and reflex amplitude (which are 10. Davatelis G, Wolpe SD, Sherry B, Dayer JM, Chicheportiche R, Cerami A:
excellent measures of opioid effect) were essentially Macrophage inflammatory protein-1: A prostaglandin-independent endogenous
pyrogen. Science 1989; 243:1066 – 8
unchanged by epidural fentanyl. As may be expected 11. Crankshaw DP, Karasawa F: A method for implementing programmed
from these data, epidural fentanyl had no detectable infusion of theiopentone and methohexitone with a simple infusion pump.
Anaesth Int Care 1989; 17:496 –9
influence on the febrile response. 12. Scott JC, Ponganis KV, Stanski DR: EEG quantification of narcotic effect:
We have previously demonstrated that opioid-induced The comparative pharmacodynamics of fentanyl and alfentanil. ANESTHESIOLOGY
1985; 62:234 – 41
inhibition of fever is mediated centrally, rather than by a 13. Gourlay GK, Kowalski SR, Plummer JL, Cousins MJ, Armstrong PJ: Fentanyl
reduction in peripheral concentrations of pyrogens, blood concentration—Analgesic response relationships in the treatment of post-
operative pain. Anesth Analg 1988; 67:329 –37
when opioids are given after IL-2.9 Our current results 14. Lehmann K: Pharmacokinetics of opioid analgesics, Patient Controlled
confirm this observation and extend it to the case in Analgesia. Edited by Harmer M, Rosen M, Vickers MD. Oxford, Blackwell Scien-
tific, 1985, pp 18 –29
which opioids are given before induction of fever. Epi- 15. Sessler DI, Schroeder M: Heat loss in humans covered with cotton hospital
dural analgesia did not inhibit fever and, not surprisingly, blankets. Anesth Analg 1993; 77:73–7
16. Belani KB, Sessler DI, Larson M, Lopez M, Washington DE, Ozaki M,
also failed to reduce circulating pyrogen concentrations. McGuire J, Merrifield B, Hynson J, Schroeder M: The pupillary light reflex: Effects
We evaluated volunteers rather than patients. Re- of anesthetics and hyperthermia. ANESTHESIOLOGY 1993; 79:23–7
17. Kurz A, Ikeda T, Sessler DI, Larson M, Bjorksten AR, Dechert M, Chris-
sponses under clinical circumstances are likely to differ tensen R: Meperidine decreases the shivering threshold twice as much as the
somewhat. Nonetheless, it remains likely that the cause vasoconstriction threshold. ANESTHESIOLOGY 1997; 86:1046 –54
18. Bjorkman S, Stanski DR: Simultaneous determination of fentanyl and alfen-
we propose for hyperthermia associated with epidural tanil in rat tissues by capillary column gas chromatography. J Chromatog 1988;
analgesia applies widely. Our theory does not exclude 433:95–104
19. Selinger K, Lanzo C, Sekut A: Determination of remifentanil in human and
other causes of hyperthermia during epidural analgesia; dog blood by HPLC with UV detection. J Pharm Biomed Anal 1994; 12:243– 8
it therefore remains likely that other yet-to-be-identified 20. de Waal Malefyt R, Haanen J, Spits H, Roncarolo MG, te Velde A, Figdor C,
Johnson K, Kastelein R, Yssel H, de Vries JE: Interleukin 10 (IL-10) and viral IL-10
mechanisms also contribute. strongly reduce antigen-specific human T cell proliferation by diminishing the
In summary, the conventional assumption is that hy- antigen-presenting capacity of monocytes via downregulation of class II major
histocompatibility complex expression. J Exp Med 1991; 174:915–24
perthermia during epidural analgesia is caused by the 21. de Waal Malefyt R, Abrams J, Bennett B, Figdor CG, de Vries JE: Interleukin
technique. We tested an alternative theory that fever in 10(IL-10) inhibits cytokine synthesis by human monocytes: An autoregulatory
role of IL-10 produced by monocytes. J Exp Med 1991; 174:1209 –20
humans is manifested normally during epidural analge- 22. Fiorentino DF, Zlotnik A, Mosmann TR, Howard M, O’Garra A: IL-10
sia, but suppressed by low-dose opioid. Intravenous fen- inhibits cytokine production by activated macrophages. J Immunol 1991; 147:
3815–22
tanyl halved the febrile response to pyrogen, whereas 23. Nava F, Calapai G, Facciola G, Cuzzocrea S, Marciano MC, De Sarro A,
epidural ropivacaine and epidural ropivacaine–fentanyl Caputi AP: Effects of interleukin-10 on water intake, locomotory activity, and
rectal temperature in rat treated with endotoxin. Int J Immunopharmacol 1997;
did not inhibit fever. These data support our proposed 19:31– 8
mechanism for hyperthermia during epidural analgesia 24. Churgay CA, Smith MA, Blok B: Maternal fever during labor—what does it
mean? J Am Board Fam Pract 1994; 7:14 –24
and suggest that hyperthermia during epidural analgesia 25. Dashe JS, Rogers BB, McIntire DD, Leveno KJ: Epidural analgesia and
should thus be taken seriously and not considered a intrapartum fever: placental findings. Obstet Gynecol 1999; 93:341– 4
26. Scott DA, Blake D, Buckland M, Etches R, Halliwell R, Marsland C, Mer-
benign complication of the anesthetic technique per se. ridew G, Murphy D, Paech M, Schug SA, Turner G, Walker S, Huizar K, Gustafs-
son U: A comparison of epidural ropivacaine infusion alone and in combination
with 1, 2, and 4 microg/mL fentanyl for seventy-two hours of postoperative
analgesia after major abdominal surgery. Anesth Analg 1999; 88:857– 64
27. Liu SS, Moore JM, Luo AM, Trautman WJ, Carpenter RL: Comparison of
References three solutions of ropivacaine/fentanyl for postoperative patient-controlled epi-
dural analgesia. ANESTHESIOLOGY 1999; 90:727–33
1. Fusi L, Maresh MJA, Steer PJ, Beard RW: Maternal pyrexia associated with 28. Baxter AD, Laganiere S, Samson B, Stewart J, Hull K, Goernert L: A
the use of epidural analgesia in labour. Lancet 1989; 1:1250 –2 comparison of lumbar epidural and intravenous fentanyl infusions for post-
2. Philip J, Alexander JM, Sharma SK, Leveno KJ, McIntire DD, Wiley J: thoracotomy analgesia. Can Anaesth 1994; 41:184 –91
Anesthesiology, V 94, No 2, Feb 2001